AN ASSESSMENT OF EFFICIENCY OF HEAVY METALS REMOVAL BY A
CONSTRUCTED WETLAND AT EGERTON UNIVERSITY, KENYA
MWANYIKA FORTINAT TEREWE
A Thesis Submitted to the Graduate School in Partial Fulfilment of the Requirement for
the Award of the Degree of Master of Science in Environmental Science of Egerton
University
Declaration
I hereby declare that this thesis is my original work and has not been submitted or
presented for examination in any other institution of learning to the best of my knowledge.
Signed: …………………………... Date………………………………..
Recommendation
We wish to confirm that this thesis was done under our supervision and has our approval
to be presented for examination as per the Egerton University regulations.
Signed: …………………………… Date:………………………………...
Department of Environmental Science
Department of Biological Sciences
© 2016 Mwanyika Fortinat Terewe
No part of this work may be reproduced, stored in a retrieval system, or transmitted by any
means, mechanical, photocopying, electronic process, recording, or otherwise copied for public
or private use without the prior written permission from the author or Egerton University.
All rights reserved.
iv
DEDICATION
This work is dedicated to my wife Rose and our children Chao, Andrew and Caleb for their
encouragement, patience and prayers during my studies.
v
ACKNOWLEDGEMENTS
First, and above all, I praise God for the gift of life and his grace during this research work. I
would like to acknowledge and thank Egerton University for allowing me to pursue my Master
of Science degree in Environmental Science; and Chemistry Department for accommodating me
while doing my research work. My special thanks go to my supervisors, Dr. George M. Ogendi
and Prof. Julius Kipkemboi, for their guidance throughout the proposal development, research
and thesis write-up; as well as their critiques and useful ideas that significantly improved my
thesis. I also greatly appreciate the assistance I received from Prof. S. T. Kariuki of Biological
Sciences Department and Mr. Caleb Otieno of the Department of Crops, Horticulture and Soils
in identifying the wetland plants. I acknowledge the admirable support of Mr. Charles Wambugu
of Mathematics Department in data analysis. I would like to express my sincere gratitude to Mr.
Leonard Kibet of the Department of Civil and Environmental Engineering, my friends Samuel
Kariuki and Paul Kamau for their technical support together with all who assisted me during my
research work. In a special way, I appreciate my family for their love, encouragement and
support.
vi
ABSTRACT
An assessment of the efficiency of heavy metals removal by a constructed wetland system at
Egerton University, Kenya, was conducted between August 2013 and January 2014. The aim of
the study was to evaluate the physico-chemical characteristics of wastewater and investigate
heavy metals retention in the wetland. Water samples were collected monthly in plastic bottles at
the inlet, along the wetland, and at the outlet; sediment samples were collected from the gravel
bed and the three wetland cells using a core sampler. Whole plants were randomly collected and
pooled together to form composite samples for each species in every site. In every sampling
occasion; temperature, pH, electrical conductivity (EC) and Dissolved Oxygen (DO) of water
samples were measured in situ. In the laboratory the samples were processed and the
concentrations of metals; lead (Pb), cadmium (Cd), copper (Cu) and zinc (Zn) determined using
Atomic Absorption Spectrophotometry (AAS). Minitab software was used to determine spatial
variations of heavy metals concentrations and physico-chemical characteristics of water using
Analysis of Variance (ANOVA). Further, correlation analyses were performed to establish
relationships between the physico-chemical parameters and the retention of metals in the
wetland. The study results showed significant variations in temperature and conductivity across
the wetland (p < 0.05). On the contrary, there was no significant difference in DO across the
wetland. Influent levels for lead, copper and zinc were 1.25 ± 0.75 mg/L, 1.09 ± 0.49 mg/L and
0.15 ± 0.11 mg/L respectively while the level of these metals in the effluent were 0.07 ± 0.07
mg/L and 0.32 ± 0.11 mg/L for lead and copper, respectively with zinc being below detection
limit. Removal efficiencies of 94%, 70% and 100% for lead, copper and zinc respectively were
observed. There was a significant negative correlation between zinc in sediment and pH in water
(r = - 0.55), and a moderate positive correlation between copper in plants and pH in water (r =
0.47). These findings indicate that the constructed wetland is efficient in removing heavy metals
from the wastewater. The study recommends that the wetland should be rehabilitated to enhance
and sustain its function of removing heavy metals from wastewater in order to safeguard human
and environmental health.
DEFINITION OF TERMS......................................................................................................... xv
1.3 Objectives ......................................................................................................................... 3
CHAPTER TWO .......................................................................................................................... 5
LITERATURE REVIEW ............................................................................................................ 5
viii
2.1.2 Heavy metal toxicity to humans and other organisms ............................................. 6
2.1.3 Lead pollution and toxicity ....................................................................................... 7
2.1.4 Zinc pollution and toxicity ........................................................................................ 7
2.1.5 Cadmium pollution and toxicity ............................................................................... 8
2.1.6 Copper pollution and toxicity ................................................................................... 9
2.1.7 Potential sources of heavy metals at Egerton University.......................................... 9
2.2 Wastewater treatment options ........................................................................................ 10
2.2.1 Conventional waste stabilization Ponds .................................................................. 10
2.3 Use of wetlands for wastewater treatment ..................................................................... 11
2.3.1 Constructed Wetlands ............................................................................................. 11
2.3.2 Use of constructed wetlands for wastewater treatment in Kenya ........................... 12
2.3.3 Types of constructed wetlands ................................................................................ 12
2.3.4 The role of vegetation in constructed wetlands ...................................................... 13
2.3.5 Macrophytes in the Egerton University Constructed Wetland ............................... 14
2.4 Heavy metals removal in wetland systems .................................................................... 17
2.5 NEMA Standards for effluent discharge into the environment...................................... 18
2.6 Conceptual framework ................................................................................................... 18
3.2 Sampling points .............................................................................................................. 23
3.3 Sample collection ........................................................................................................... 26
ix
3.4.1 Heavy metals analysis in wastewater samples ........................................................ 26
3.4.2 Heavy metals analysis in plant samples .................................................................. 27
3.4.3 Heavy metals analysis in sediment samples ........................................................... 27
3.5 Data analysis .................................................................................................................. 28
4.2 Heavy metals concentration in the wetland................................................................... 30
4.2.1 Heavy metals concentration in water samples ....................................................... 30
4.2.2 Heavy metals concentration in sediment samples ................................................. 31
4.2.3 Heavy metals concentration in plant samples ........................................................ 32
4.3: Relationship between physico-chemical parameters and heavy metals in the wetland. .... 33
4.4 Heavy metals removal efficiency in the wetland ........................................................... 37
4.5 Discussion ...................................................................................................................... 38
4.5.2 Heavy metals concentration in water samples ....................................................... 39
4.5.3 Heavy metals concentration in sediment samples ................................................. 40
4.5.4 Heavy metals concentration in plant samples ......................................................... 40
4.5.5 Relationship between physico-chemical parameters of water and heavy metals
concentrations in the wetland................................................................................................ 41
Plate 6: Wetland cell 1 .................................................................................................................. 24
Plate 7 : Wetland cell 2 ................................................................................................................. 24
Plate 8 : Wetland cell 3 ................................................................................................................. 25
Plate 9: River Njoro at the wastewater discharge point ............................................................... 25
xii
Figure 2: Map of the study area ................................................................................................... 22
Figure 3: Variation of metal concentrations along the wetland profile ........................................ 31
xiii
LIST OF TABLES
Table 1: Partial NEMA standards for effluent discharge into the environment ........................... 18
Table 2: Main Components of the Constructed Wetland and their sizes ..................................... 21
Table 3: Mean values and ranges of physico-chemical parameters .............................................. 29
Table 4: Mean values of selected metals concentration in water .................................................. 30
Table 5: Mean values of selected metals concentrations in sediment ......................................... 32
Table 6: Mean values of metals concentration in plants ............................................................... 32
Table 7: Correlation coefficients between temperature and heavy metals . ................................. 33
Table 8: Correlation coefficients between DO and heavy metals ................................................ 34
Table 9: Correlation coefficients between pH and heavy metals ................................................. 35
Table 10: Correlation coefficients between EC and heavy metals ............................................... 36
Table 11: Metal removal efficiency in the wetland ...................................................................... 37
xiv
ATSDR Agency for Toxic Substances and Disease Registry
BOD Biochemical Oxygen Demand
DO Dissolved Oxygen
GEF - SGP Global Environment Facility - Small Grants Programme
NEMA National Environment Management Authority
NGO Non-Governmental Organization
SDGs Sustainable Development Goals
USDOD United States Department Of Defence
WC1 Wetland Cell 1
WC2 Wetland Cell 2
WC3 Wetland Cell 3
WHO World Health Organization
WSPs Waste stabilization Ponds
xv
DEFINITION OF TERMS
Heavy metal: a general term given to the group of metals and metalloids with a specific density
greater than 5g/cm3.
Wetlands: lands transitional between terrestrial and aquatic systems where the water table is
usually at/or near the surface or the land is covered by shallow water.
Wastewater: used water with dissolved or suspended solids, discharged from homes,
commercial establishments, farms, and industries.
Constructed wetland: engineered systems designed and constructed to utilize the natural
processes involving wetland vegetation, soils, and the associated microbial assemblages to assist
in treating wastewaters.
Phytoremediation: the direct use of living green plants for in situ removal, degradation, or
containment of contaminants in soils, sludge, sediments, surface water and groundwater.
Sorption: the process in which one substance takes up or holds another (by either absorption or
adsorption).
Absorption: soaking up; a process in which one substance permeates another; a fluid permeates
or is dissolved by a liquid or solid).
Adsorption: surface assimilation (the adhesion of atoms, ions or molecules from a gas, liquid or
solid to a surface).
Chemisorption: adsorption in which the adsorbed substance is held by chemical bonds.
Biosorption: a physiochemical process that occurs naturally in certain biomass which allows it
to passively concentrate and bind contaminants onto its cellular structure.
Precipitation: the formation of a solid in a solution or inside another solid during a chemical
reaction or by diffusion in a solid.
Co-precipitation: the carrying down by a precipitate of substances which are normally soluble
under the condition of precipitation.
Physico-chemical: joint action of both physical and chemical properties.
xvi
Biomagnification: the concentration of toxins in an organism as a result of ingesting other plants
or animals in which the toxins are more widely disbursed.
Bioaccumulation: accumulation of substances, i.e. pesticides or heavy metals in an organism.
Redox potential: a measure of the tendency of a chemical species to acquire electrons and
thereby be reduced.
Hydraulic loading rate (HLR): volume of water entering the wetland divided by the surface
area of the wetland.
Hydraulic retention time (HRT): the average time that water remains in the wetland.
Metal removal efficiency: the ratio of the concentration of a metal at the outlet of the wetland to
its influent concentration.
1.1 Background to the study
The world’s freshwater resource is scarce and vulnerable to pollution by waste streams
generated by human activities. The decline of water quality across the world is constraining
global freshwater supplies and millions of people lack access to adequate supply of drinking
water (WHO/UNICEF, 2012). Hence most of the problems that humanity face in the twenty-first
century are linked to water quantity and quality issues (UNESCO, 2009).
The major activities responsible for increase in water pollution are industrialization,
urbanization and rapid growth in population. Pollutants, whether organic or inorganic, severely
impact human health and the stability of natural ecosystems. It is projected that by the year 2050,
untreated wastewater could reduce the world’s freshwater resources by as much as 18,000 km3
annually (Bao and Kuyama, 2013).
The practice of discharging wastewater into watercourses and natural wetlands was a
common practice as a means of disposal in the past (McEldowney et al., 1998). This method is
effective where discharges do not exceed the capacity of the receiving system to assimilate,
absorb or detoxify contaminants present (Bayes et al., 1989). The major chemical pollutants in
wastewater include ionic forms of nitrogen and phosphorus, heavy metals, detergents, pesticides,
and hydrocarbons (Larsdotter, 2006). Unlike pollution associated with major plant nutrients in
aquatic environment such as nitrogen and phosphorus, which can be assimilated to usable
biomass at different trophic levels, heavy metals tend to accumulate and magnify at higher levels
of the food chain. This is attributed to their non-biodegradability and persistence in the
environment.
Heavy metal contamination can have devastating effects on human health and the
ecological integrity of the receiving environment (Farombi et al., 2007). Although living
organisms require varying amounts of heavy metals such as Iron, cobalt, copper, manganese,
molybdenum and zinc, excessive levels can have deleterious effect on organisms (Rainbow,
2007). Zinc, for example, when present in high levels inhibit many plant metabolic functions
resulting to retarded growth and senescence (Ebbs and Kochian, 1997), while high concentration
2
of copper may severely damage gills, adversely affect the liver and kidneys of fish or cause some
neurological damage (Flemming and Trevors, 1989).
Removal of heavy metals from wastewater can be accomplished through various
treatment options, including chemical precipitation, coagulation, complexation, activated carbon
adsorption, ion exchange and reverse osmosis. Although chemical precipitation has been
extensively practiced in the removal of heavy metals from wastewaters (McEldowney et al.,
1998), this technique is costly, requires continued supervision and sometimes present operational
challenges (Upadhyay, 2004). Thus, effluents released from wastewater treatment plants may
contain various heavy metals (Mapanda et al., 2005).
The use of natural and artificial (constructed) wetlands for wastewater treatment has been
proposed as an intermediate technological solution for handling wastewater (McEldowney et al.,
1998). These systems are economically attractive and relatively energy-efficient for wastewater
treatment, especially for isolated populations, yet with capacity to improve aesthetic value of an
area. The basic concept of these systems is deceptively simple and many constructed wetlands
have not performed as expected.
Improvement in water quality in a wetland system may be assessed by monitoring and
comparing a range of water-quality parameters at the inflow and outflow. Some of the
parameters commonly used for monitoring wetland performance include dissolved oxygen
content, biochemical oxygen demand, total phosphorus, total nitrogen, faecal coliform, pH,
suspended solids, electrical conductivity, and heavy metal content. Corrective measures are then
applied when results show that the system is not working according to the set objectives
(Kayombo et al., 2001).
The wastewater stabilization ponds at Egerton University were constructed several
decades ago when the institution was an agricultural college with a few hundreds of students and
staff. Upon upgrading to a fully-fledged University in 1987, the population on campus increased
steadily and so did the volume, as well as the complexity of wastewater. In 2007, a wetland was
constructed to polish the effluent from the waste stabilization ponds before discharging it into
River Njoro.
1.2 Statement of the problem
The discharge of treated wastewater into River Njoro has been an issue of concern to the
general public and other stakeholders. The river not only serves as a source of water for
downstream communities but also provides the bulk of the total freshwater inflow into Lake
Nakuru – a wetland of international importance. Previous studies have reported the working
status of the WSPs at Egerton University as only moderate, with respect to nutrients, pathogens
and traces of heavy metals in the effluent. Heavy metals have been detected in water, plants, fish
and sediments in Lake Nakuru. Heavy metals persist in the environment, bio-accumulate in
organisms and bio-magnify along the food chain. There have been episodes of flamingo deaths
in Lake Nakuru and heavy metals were found accumulated in the organs of their carcases. This
led to the listing of heavy metal toxicity as one of the possible causes for their death. Therefore,
water contaminated with heavy metals may adversely affect human health for the River Njoro
riparian populations and the Lake Nakuru biodiversity.
1.3 Objectives
1.3.1 General Objective
To assess the efficiency of Egerton University constructed wetland in removing heavy
metals from wastewater.
1.3.2 Specific objectives
conductivity and dissolved oxygen) of water across the wetland profile.
2. To determine the concentrations of lead, copper, cadmium and zinc in water, plants
and sediments in the wetland.
3. To assess the effect of selected physico-chemical parameters on retention of heavy
metals in the wetland.
H01: The physico-chemical parameters of water have no significant variation along the
wetland profile.
H02: The concentrations of lead, copper, cadmium and zinc in water, plants and sediments
have no significant difference across the wetland profile.
H03: There is no significant relationship between selected physico-chemical parameters of
water and the retention of heavy metals in the wetland.
1.5 Justification of the study
Egerton University is obligated to protect human health and environmental integrity as
well as comply with regulations and related standards for effluent discharge into the
environment. There is also the need to implement sustainable environmental management
programmes that meet international standards (ISO 14001:2015) and actively participate towards
the achievement of Vision 2030 of Kenya and the SDGs (Goal 6), (UN, 2015). It is also a
moral/ethical obligation for the University to ensure that the water quality of River Njoro does
not change to the detriment of downstream users. There is also need to protect the flora and
fauna in Lake Nakuru, a Ramsar site, from the effects of heavy metals pollution. Effectively
treated wastewater has great potential for re-use by humans and livestock as well as supplement
existing freshwater supplies. Data obtained in this study will provide valuable information on the
wetland and the entire wastewater treatment system at Egerton University.
1.6 Scope of the study
The study covered Egerton University constructed wetland only and investigated selected
physico-chemical parameters of wastewater (pH, temperature, Electrical conductivity and
dissolved oxygen) and the content of selected heavy metals (Pb, Cu, Cd and Zn) in water,
dominant macrophytes (P. stratiotes, C. alopecuroides, S. lacustris, and E. crassipes) and
sediments along the entire wetland profile. The study was conducted over a period of six months
(August 2013 to January 2014).
5
2.1 Global water pollution
Water pollution is a major cause of the water crisis being experienced globally. The ever
increasing population, industrialization, food production practices, improved living standards and
poor water management approaches are responsible for this scenario. The water degradation
risks impacts directly into our social and economic activities (UN, 2012). Most of the significant
forms of water pollution worldwide are caused by inadequate sanitation infrastructure leading to
contamination of watercourses (UNICEF /WHO, 2008). It is estimated that about 2 million tons
of sewage and industrial waste are discharged daily into the world’s water bodies. Thus, many
important water bodies, which provide water for domestic use and irrigation, are displaying
undesirable levels of potentially toxic substances (RSC, 2010).
Poor water quality is a risk to human and environmental health, escalates treatment costs
and reduces availability of safe water (Palaniappan et al., 2010). About 900 million people
globally lack access to safe drinking water (UNDESA, 2009), while almost 2.6 billion do not
have adequate sanitation (WHO/UNICEF, 2010). Consequently, polluted water is a major cause
of disease and death in the developing world where about 80% of all diseases and over one third
of deaths can be linked to use of polluted water (Pandey, 2006). In Kenya, about 43% of the
population have no access to clean water and this severely affects many lives with many illnesses
(Marshall, 2011). Worldwide, water related diseases are the number one killer of children below
the age of five years (WHO, 2002). Although microbiological hazards form the largest
contributor to waterborne diseases in the world, chemical pollutants can bring serious health
problems irrespective of whether the source of chemicals are natural or anthropogenic (Terrence
et al., 2007).
2.1.1 Heavy metals occurrence in the environment
Environmental contamination with heavy metals is a growing problem globally and has
worryingly increased in the last five decades due to the innumerable uses of metals in industrial
processes and products (El-Safty, 2014). Metals contamination is one of the oldest environmental
problems and still remains a serious health issue today. The widespread use of metals, the
6
legacies of the past contamination and novel technologies continue to pose ecological risks to
aquatic environments globally (Luoma and Rainbow, 2008).
Heavy metals are commonly defined as those having a specific density greater than 5
g/cm3 and adversely affect the environment and living organisms (Järup, 2003). This group of
metals are natural constituents of the earth crust and can be released into the environment
through natural processes (e.g. weathering and volcanic eruptions) and various anthropogenic
activities including sewage and storm water discharges, landfills, mining and smelting,
electroplating, laboratories, tanning, battery manufacture, electronic waste disposal and many
others (Morais et al., 2012). The most common heavy metals found in wastewater include
arsenic, cadmium, chromium, copper, lead, nickel and zinc (Lambert et al., 2000). Heavy metals
are significant environmental pollutants and their toxicity present a problem of increasing
importance for ecological, evolutionary, nutritional and environmental concerns (Jaishankar et
al., 2014). Heavy metals cannot be degraded to harmless by-products by any biological, physical
or chemical means but can be transformed from one oxidation state or organic complex to
another.
2.1.2 Heavy metal toxicity to humans and other organisms
Generally, toxicity of heavy metals is attributed to the high affinity for ligands like
sulphur and nitrogen, by which evolution exploits heavy metals as essential components of
metabolism. This makes all heavy metals potentially toxic by binding to proteins or other
molecules, consequently inhibiting them from functioning in their typical metabolic role (Louma
and Rainbow, 2008). Heavy metals interfere with numerous enzymes and hormones function and
disrupt many metabolic and physiological processes in the body (Mukke and Chinte, 2012).
They compete with essential trace elements for binding sites on transport and storage proteins,
metallo-enzymes and receptors. This leads to abnormalities in the metabolism of carbohydrate,
protein/amino acids, lipids, neurotransmitters, hormones, and increase vulnerability to infections
(El-Safty et al., 2009). Heavy metals are particularly toxic when exposure occur early in life
because they affect brain development and other organ systems at very low levels that are
presumed to be safe for adults (ATSDR, 2005).
Among the best known heavy metal contamination are the episodes of the 1950s and
1960s that occurred around Minamata Bay, Japan, where about 3000 people, including many
7
infants, suffered birth defects and severe nervous system damage from consuming shellfish and
fish contaminated with mercury (Luoma and Rainbow, 2008). Another major occurrence of
mercury poisoning happened in Northern Iran in 1972, where farmers consumed wheat seeds
treated with mercurial fungicides instead of planting them (Freedman, 1989). Although human
health episodes raised awareness of threats from heavy metal contamination, ecological damage
is likely to be more persistent. In aquatic environments, heavy metals accumulate in upper
sediment. The metals can be released by biological and geochemical mechanisms and become
toxic to sediment-dwelling organisms and fish, leading to death, reduced growth, or reduced
reproduction leading to lower species diversity (Praveena et al., 2007).
2.1.3 Lead pollution and toxicity
Lead is the commonest of the heavy metals, accounting for 13 mg/kg of Earth’s
crust (WHO, 2011). Lead has been widely used in construction, piping, in pigments, solders,
shields for protection against radioactive materials, as an insecticide (lead arsenate), ceramic
glazing, crystal tableware among other uses (Wright and Welbourne, 2002). Although most of
these uses have been discontinued, their legacies of lead contamination in the environment still
remain.
Lead is one of the most significant toxin of the heavy metals. Its toxicity causes
significant changes in many biological processes. It is known to replace bivalent cations such as
Ca2+, Mg2+, Fe2+ and monovalent cations like Na+ leading to interference in the normal
metabolism of the cell. Lead bio-accumulates in the body and this may lead to decrease in
haemoglobin production, kidney, joint, reproductive and cardiovascular systems disorders as
well as long-term damage to the central nervous system (Galadima and Garba, 2012). High lead
concentration in a plant enhances the production of reactive oxygen species (ROS), causing lipid
membrane damage which eventually interferes with chlorophyll and photosynthetic processes,
hence reducing the overall growth of the plant (Srivastava et al., 2015).
2.1.4 Zinc pollution and toxicity
Zinc, the 25th most abundant element in the earth’s crust (78mg/kg), occurs only in the
divalent state and does not occur as a metal in nature (Alloway, 2003). Zinc is used in dry
batteries, in construction, galvanizing, making coins, alloys (e.g. brass and bronze), solder,
8
paints, rubber products, cosmetics, printing ink, soap among others (Wright and Welbourne,
2002). Zinc is also used in dental, medical, and many household applications. Organo-Zn
compounds are used as fungicides, topical antibiotics, and lubricants (Simon-Hettich et al.,
2001). It is a heavy metal of much interest since it is an essential element in living organisms as
well as a potential contaminant in the environment. It is an essential plant micronutrient and a
constituent of many metalloenzymes and transcription factors which are involved in many
cellular processes such as gene expression, signal transduction, transcription, and replication
(Berg and Shi, 1996).
Zinc is considered to be relatively non-toxic, especially when taken orally. However,
excess amount can cause system dysfunctions that result in impairment of growth and
reproduction (Nolan, 2003). The clinical signs of zinc poisoning have been reported as vomiting,
diarrhoea, bloody urine, jaundice, liver failure, kidney failure and anaemia (Fosmire, 1990).
Although the risk of excess Zn in humans from environmental sources is apparently low, zinc
from anthropogenic sources spreads widely in the environment and adversely affect other
organisims. In fish, for example, zinc causes decrease in oxygen consumption because it
damages gills and also reduces growth and fertility (Pandey, 2006).
2.1.5 Cadmium pollution and toxicity
Cadmium is chemically similar to zinc and occurs naturally with zinc and lead sulphide.
Cadmium forms organic complexes with organic ligands such as humic and fulvic substances. It
is obtained as a by-product when refining zinc and other metals, especially copper and lead
(Wright and Welbourne, 2002). Rocks mined to produce phosphate fertilizers contain varying
amounts of cadmium. Occasionally the produced phosphate fertilizers have cadmium
concentrations of up to 300 mg/kg hence the high cadmium content in agricultural soils (Grant
and Sheppard, 2008). Sewage contains cadmium from human excretion, runoff from agricultural
fields, industrial effluents and disposal of domestic waste containing zinc (Wright and
Welbourne, 2002). Cadmium has many industrial uses including nickel-cadmium batteries,
electroplating, pigments, plastic plasticizers, welding electrodes and alloys.
A very wide range of concentrations have been reported in foodstuffs from various parts
of the world. Cadmium is mainly found in fruits and vegetables due to its high rate of soil-to-
plant transfer (Satarug et al., 2011). However, there have been a few cases of general population
9
toxicity as a result of chronic exposure to cadmium in contaminated food and water. In the
1940s, Japanese mining activities contaminated the Jinzu River with cadmium and traces of other
toxic metals. Rice crop grown by the river banks downstream accumulated cadmium and when
local communities consumed the contaminated rice they developed itai-itai disease and renal
abnormalities, including proteinuria and glucosuria (Nogawa et al., 2004). In fish, cadmium
accumulates in the muscle, liver, gills and bones where it exerts deleterious effects in terms of
nephrotoxicity, cytotoxicity, genotoxicity and immunotoxicity (Lippmann, 2000).
2.1.6 Copper pollution and toxicity
Copper is present in the earth’s crust at a concentration of about 50 mg/kg (Emsley,
2003). The element has various roles in biological electron transfer and oxygen transportation,
processes that exploit the easy inter-conversion of Cu (I) and Cu (II) (Lippard and Berg, 1994).
The major activities that lead to mobilization of copper into the environment are extraction from
its ores, agriculture and waste disposal. Humans are exposed to copper through inhalation of
particulate copper, drinking water and eating food contaminated with copper. However, the
toxicity of copper to humans is relatively low compared to other metals such as mercury,
cadmium, lead, and chromium. When a person is exposed to copper levels above the essential
levels needed for good health, the liver and kidneys produce metallothionein (low molecular
weight) which binds with the copper to form a water-soluble complex which is then excreted
(Bradl, 2005).
Chronic effects of copper exposure can cause liver and kidney damage. However,
mammals have effective mechanisms of regulating copper and are hence generally protected
from excess dietary copper levels (DES, 2005). High levels of free copper have been detected in
people with Alzheimer’s disease (Brewer, 2010). In this disease, copper and zinc are known to
bind to the amyloid beta proteins (Faller, 2009). It is thought that the bound form of the protein
mediates the production of reactive oxygen species (free radicals) in the brain (Hureau and
Faller, 2009).
2.1.7 Potential sources of heavy metals at Egerton University
Egerton University has expanded rapidly in the last decade, and as described by
Alshuwaikhat and Abubakar (2008), universities these days are ‘small cities’ because of their
10
populations, and the various complex activities that take place in campuses. Thus wastewater
generated from these institutions contain a wide range of contaminants that may include heavy
metals, detergents, pesticides, pathogens, hydrocarbons, pharmaceuticals etc., which can prove
difficult to treat effectively within existing systems.
Probable sources of heavy metals in the wastewater stream at Egerton University may
include, but not limited to, teaching laboratories, agricultural demonstration fields and storm
water. Storm water carries along with it run-off from paved areas, agricultural fields, cow sheds
and lawns which may contain leached metals from carelessly disposed electronic waste,
fungicides and soil. Some amounts of heavy metals are excreted through faeces and urine of
human beings and animals. According to Davies and Nightingale, (1975), zinc is primarily
excreted via the gastrointestinal tract and eliminated in the faeces after oral exposure in human
beings; approximately 70-80% of an ingested dose is excreted in the faeces. Corrosive liquid
waste from laboratories may cause solubilisation of heavy metals in the waste stream. The waste
stabilization ponds may be a secondary source of heavy metals. Depending on the physico-
chemical conditions of the ponds, the metals buried in the sludge may re-enter the water column.
2.2 Wastewater treatment options
Wastewater is 99.9% water and 0.1% solid matter, of which about half is suspended
while the other half is in dissolved form (FAO, 2004). The major chemical pollutants in domestic
and institutional wastewater include nitrogen, phosphorus, heavy metals, detergents, pesticides,
and hydrocarbons (Larsdotter, 2006). Wastewater treatment is necessary in order to produce a
disposable effluent which will not pollute and injure the receiving environment (Khopkar, 2004).
There are various processes available for wastewater treatment systems with inherent advantages
and disadvantages. Treatment can be done in decentralized systems (i.e. in septic tanks, bio
filters or anaerobic treatment systems), or collection via a sewer network to a centralized system
(e.g. a municipal treatment plant). Wastewater treatment can involve physical, chemical or
biological processes or a combination of these processes depending on the required effluent
standards (EPA, 1997).
2.2.1 Conventional waste stabilization Ponds
Waste stabilization ponds are popular for treating wastewater in tropical and subtropical
regions because of the abundant sunlight and high ambient temperatures (Hodgson, 2007). They
11
are man-made earthen basins and may include one or more series of anaerobic, facultative and
maturation ponds depending on the quality of effluent required. Anaerobic and facultative ponds
are designed for removal of Biochemical Oxygen Demand (BOD), and maturation ponds for
pathogen removal, although some BOD removal also occurs in maturation ponds and some
pathogen removal in anaerobic and facultative ponds (Mara, 1997). Wastewater stabilization
ponds have been used to treat a variety of wastewaters, from domestic wastewater to complex
industrial waters, and they operate under a wide range of weather conditions. They can be used
alone or in combination with other treatment processes (USDOD, 2004).
WSPs are popular for municipal sewage purification, especially in developing countries,
due to their cost-effectiveness and high potential of removing different pollutants. In addition,
they are easy to operate and require minimal maintenance. They have been proved to be reliable,
economic, flexible and adaptable, and able to meet the highest effluent standards (Kambole,
2003). Prior to treatment in the WSPs, wastewater is first subjected to preliminary treatment
(screening and grit removal) to remove large and heavy solids.
2.3 Use of wetlands for wastewater treatment
Generally, wetlands are lands transitional between terrestrial and aquatic systems where
the water table is usually at/or near the surface or the land is covered by shallow water.
Saturation with water is the main factor determining the nature of soil development and the types
of plant and animal communities living in the soil and on its surface (Cowardin et al., 1979).
Wetlands, both natural or constructed, are effective in lowering biochemical oxygen demand
(BOD), suspended solids, nitrogen and phosphorus, as well as reducing concentrations of metals,
organics and pathogens (Kadlec and Knight, 1996). Early history of wastewater treatment say
that wastewater was collected and discharged into natural watercourses or wetlands devoid of
any form of treatment (Rousseau, 2005). However, limits of assimilative capacity of natural
receiving waters compelled the necessity to develop technologies that treat wastewater streams
before they are discharged (Kadlec and Knight, 1996). It is the discovery of the purifying ability
of natural wetlands that led to the development of constructed wetlands.
2.3.1 Constructed Wetlands
Constructed wetlands (CWs) are engineered systems that have been designed and
constructed to utilize the natural processes comprising wetland vegetation, soils, and the
12
systems utilize the physical, chemical, and biological processes existing in wetland ecosystems
to remove and trap pollutants. Constructed wetlands for wastewater treatment have evolved
during the last five decades into a reliable treatment technology which can be applied to all types
of wastewater including sewage, industrial and agricultural wastewaters, landfill leachate and
storm water runoff (Vymazal, 2010). Constructed wetlands have shown high effectiveness in
removing heavy metals; both subsurface flow and surface flow systems having similar removal
capability (Reed et al., 1995).
2.3.2 Use of constructed wetlands for wastewater treatment in Kenya
The use of constructed wetlands for treatment of wastewater in Kenya is a fairly new idea and
has not been well exploited, although CWs have almost limitless potential (Raymer, 2006).
These systems have been successfully used to polish domestic wastewater (Nzengya and
Witshitemi, 2001) as well as industrial wastewater from pulp and paper effluents (Abira et al.,
2003). The splash Carnivore wetland, which was established in 1994, handles up to 80 m3 of
wastewater daily with acceptable pollutants removal efficiencies (Nyakang‘o and van Bruggen,
1999). There are other operational constructed wetlands in Kenya which include: Nandi Hills
Tea Factory CW that recycles agricultural waste and toilet effluent, The Kingfisher constructed
Wetland at Naivasha, Shimo La Tewa Prison, and Mombasa – Kenya, Mara Simba lodge and
Masai Mara National reserve.
2.3.3 Types of constructed wetlands
Constructed wetlands can be classified into two types according to the water flow mode
and the dominant aquatic plant species (Kadlec et al., 2000):
1. Surface flow (SF) or Free-Water-Surface (FWS) Constructed Wetlands: where the
water flows above ground and these can be further categorized into three subtypes
according to the dominant plant species i.e. emergent aquatic macrophytes,
floating aquatic macrophytes and submerged aquatic macrophytes.
2. Subsurface Flow (SSF) Constructed Wetlands: with the water flow below ground
and can be categorized into two subtypes; horizontal and vertical subsurface flow
according to the water current direction.
13
These types of wetland systems also differ from one another in layout, the removal efficiency of
certain pollutants, area requirements, technical complexity, applications, and costs. The free
water surface wetland technology closely resemble natural wetlands. They consist of large,
shallow lagoons that contain submerged, emergent, or floating plant species. The
microorganisms responsible for biological treatment of the wastewater form bio-films on the
stems and leaves of the plants. The name “free surface” originates from the thin free water layers
which are formed at the surface (Ulsido, 2014). These systems can be used for secondary
treatment of wastewater and removal of heavy metals, but they are commonly used as tertiary
treatment to remove nutrients to prevent eutrophication in the receiving water bodies (Gauss,
2008). Hybrid systems have also been developed and these combine different types of
constructed wetlands so as to improve treatment efficiencies. They are useful when more
stringent discharge limits for nitrogen are required and also more complex wastewaters are
treated in constructed wetlands (Vymazal, 2013).
2.3.4 The role of vegetation in constructed wetlands
Vegetation is an essential component of constructed wetlands (Lim et al., 2001). Plants
provide a large surface area for attachment and growth of microbes. The physical components of
the plants stabilize the surface of the beds, slow down the water flow thus assist in sedimentation
(Brix, 1994). However, not all wetland species are suitable for wastewater treatment because
plants for constructed wetland must be able to tolerate a combination of incessant flooding and
exposure to fairly high amounts of pollutants. According to Heers (2006), the three most
essential criteria for choosing wetland plants include: availability in the climate zone, pollutant
removal capacity (treatment efficiency) and tolerance ranges; and plant productivity and biomass
utilization options.
Plants also have the ability to remove nutrients, trace elements, and organics from the
water through biological uptake and surface adsorption. During photosynthesis, plants utilize
carbon dioxide and release oxygen. Submerged aquatic plants growing within the water column
elevate the dissolved oxygen level in the wetland surface water and deplete the dissolved carbon
dioxide, leading to an increased pH. Rooted wetland macrophytes also actively transport oxygen
from the atmosphere to the sediments thus supporting aerobic and facultative anaerobic
microorganisms in the otherwise anaerobic sediments and soils (ITRC, 2003). However, plant
14
senescence can lead to large nutrient and metal releases, compromising the treatment abilities of
a constructed wetland (Kroger et al., 2007).
2.3.5 Macrophytes in the Egerton University Constructed Wetland
A free-water surface wetland was constructed to polish pre-treated wastewater effluent
from the waste stabilization ponds. The wetland system consists of one vegetated gravel bed and
a series of three connected, vegetated wetland cells. The dominant plants in the constructed
wetland are Eichhornia crassipes, Pistia stratiotes, Cyperus alopecuroides and Scirpus lacustris.
1. Eichhornia crassipes
The Eichhornia crassipes (Water Hyacinth), (Plate1) is a member of the pickerelweed
family (Pontederiaceae). It is a free-floating vascular plant, which adapts easily to various
aquatic conditions and plays a significant role in extracting and accumulating metals from water
(Weiliao and Chang, 2004). This ability is useful in removing toxic heavy metals and trace
elements from contaminated soils and water in a process referred to as phytoremediation where
plants are used to extract, sequester and/ or detoxify (Meagher, 2000).
Plate 1: Eichhornia crassipes (Photo by Mwanyika F.T 2013)
2. Cyperus alopecuroides
Cyperus alopecuroides (Foxtail flatsedge), (Plate 2) is a perennial herbaceous plant with
a very short thick woody rhizome that grows up to 150 cm tall on shallow waters. The stems are
erect, solitary, stout and trigonous with width of about 2-5 mm. They are leafy and thickened at
base by sheaths and intravaginal shoots. Its leaves, generally shorter than the stems, are papery
15
with blades 1-12 mm wide that are flat with prominent midrib and numerous lateral veins. It has
4-7 leaf-like bracts that can grow up to 70 cm long. The plant has a stem which carries a single
inflorescence in the shape of an umbel with 5-10 primary rays, which can grow to 15 cm in
length. It has 1-8 clustered; cylindrical spikes usually 5-10 mm wide and 15-40 mm long (Brullo
and Sciandrello, 2006). Prusty et al. (2007) reported that C. alopecuroides showed capacity to
adsorb copper.
3. Pistia stratiotes
Pistia stratiotes (Water lettuce), (Plate 3) is a free floating perennial aquatic plant native
to South America. It is stoloniferous, forms colonies, and has rosettes up to 15 cm across. Leaves
are light green and velvety - hairy with obvious parallel longitudinal veins, which are slightly
broad and are widest at the apex (Ramey, 2001). It reproduces rapidly by vegetative offshoots
formed on short, brittle stolons although it is also known to reproduce by seed. Pistia stratiotes
has been included in the Global Invasive Species Database (GISD, 2005). Studies have shown
that Pistia stratiotes is a bioaccumulator and a bioindicator of various heavy metals, including
cadmium, chromium, copper and zinc (Mishra and Tripathi, 2008).
16
4. Scirpus lacustris
Scirpus lacustris (Bulrush), (Plate 4) is an emergent perennial macrophyte. It has
shallow, dense root and rhizome system that extends near the surface and the plant grows up to a
height of 2.5 – 3 m. It prefers soils that are slightly acidic or slightly alkaline and can tolerate
permanent inundation up to 0.3 – 0.4 m and is sensitive to prolonged droughts. Scirpus lacustris
shows a high adaptability to various wastewater conditions and potential to absorb, translocate
and concentrate chromium in its tissues (Heers, 2006 and Yadav et al., 2005)
Plate 4: Scirpus lacustris (Photo by Mwanyika F.T 2013)
17
2.4 Heavy metals removal in wetland systems
When heavy metals enter a constructed wetland, they are distributed between the
substrate, the water column and the vegetation (Sheoran and Sheoran, 2006). There are three
main wetland processes that are involved in removal of heavy metals; namely, binding to soils,
sediments and particulate matter, precipitation as insoluble salts, and uptake by bacteria, algae
and plants (Kadlec and Knight, 1996).
The major physico-chemical removal processes of heavy metals occurring in CWs
include (1) sedimentation and filtration, (2) sorption and (3) precipitation and co-precipitation.
Sedimentation and filtration are important physical processes allowing the removal of heavy
metals associated with particulate matter. Sedimentation has long been recognized as a principle
process in the removal of heavy metals from wastewaters (Sheoran and Sheoran, 2006). Sorption
of heavy metals on a surface may occur by processes of adsorption and precipitation. Heavy
metals may be adsorbed either electrostatically, resulting in the formation of relatively weak
complexes (physical adsorption) or chemically, resulting in the formation of strong complexes
(chemisorption). Metals are more strongly bound by chemisorption than by physical adsorption
(Evangelou, 1998). Metals that are adsorbed on the surface of the substrate can be exchanged by
other cations by the process of cation exchange. Next to the sorption and sedimentation
reactions, metals can also precipitate with (oxy-) hydroxides, sulphides, carbonates, etc. when
solubility products are exceeded (Kadlec and Knight, 1996). Stability of metals precipitated as
inorganic compounds is primarily controlled by pH. Redox potential (Eh) and pH in the wetland
affect the reduction-dissolution of Fe and Mn oxides, which in turn affect the release of heavy
metals (Nimirciag and Ajmone-Marsan, 2014).
Some wetland plants have proven to be good accumulators of heavy metals
(hyperaccumulators). Once the heavy metals have been accumulated in the plant tissues via
phytoextraction, the plants may be harvested for disposal or for incineration to retrieve the
metals from the plants. Micro-organisms also affect the behaviour of heavy metals in constructed
wetlands by their role in the biogeochemical cycles which affect metal speciation, biosorption,
reduction and methylation of heavy metals (Kosolapov et al., 2004).
Wetlands have been used to effectively purify metal-contaminated mine drainage water
(Weis and Weis, 2004), and can offer an alternative to conventional treatment of contaminated
water. A substantial reduction of 99.6% of Pb, 94.4% of Cd and 94.8% of Zn was reported in
18
constructed wetland in Sri Lanka (Jayaweera et al., 2006). Similarly, high removal efficiencies
of 80.0%, 95.1% and 100% for Pb, Cu and Zn respectively have been reported at Nandi Hills tea
estates wetland, Kenya (Gituku et al., 2015).
2.5 NEMA Standards for effluent discharge into the environment
The National Environment Management Authority (NEMA) came up with guidelines for
discharge of effluent into the environment in 2006 and may add any other parameters as may be
prescribed by the Authority from time to time. Table 1 shows the regulatory standards for some
heavy metals and selected parameters in wastewater discharged into the environment.
Table 1: Partial NEMA standards for effluent discharge into the environment
Parameter Max Allowable Limits
Aluminum 0.02 mg/L
Chromium VI 0.0 mg/L
Fluoride 1.5 mg/L
Zinc 0.5 mg/L
(Source: Kenya Gazette supplement No 68, 2006)
2.6 Conceptual framework
Metals in wastewater are present in colloidal, particulate, and dissolved phases. When
wastewater enter the wetland, the metals partition into three main compartments namely the
substrate, water column and vegetation through interaction of different variables and prevailing
geochemical conditions (Figure 1).
Intervening Varibles
Figure 1: A conceptual framework showing some of the wetland processes and factors affecting
metals retention in the wetland
There are several interconnected factors that influence heavy metals retention in a
wetland which include: pollutant loading rate, hydraulic retention time (HRT), hydraulic loading
rate (HLR), climatic conditions, temperature, pH, oxygen availability, wetland design
components – substrate, vegetation and living organisms (DeBusk, 1999). According to Lesage
et al., (2007) there are four main mechanisms that affect metal retention in wetlands: (1)
adsorption to fine sediments and organic matter, (2) precipitation as insoluble salts (such as
sulphides and oxyhydroxides), (3) absorption and induced changes in biogeochemical cycles by
plants and bacteria (Kaldec and Knight, 1996), and (4) deposition of suspended solids because of
low flow rates. A decrease in pH will increase the competition between metals and hydrogen
ions for binding sites and may dissolve metal complexes, releasing free metal ions into the water
column. An increase in pH is generally accompanied by a decrease in the solubility of many
toxic heavy metals in water (Ebrahimpour and Mushrifah, 2008). Temperature influences most
of the processes taking place in the wetland. Roots of wetland plants provide surface for plaque
formation which limits the movement of metals in the water column for a period of time until
there is a change in physicochemical conditions which can release the metals back into the water
column (Hansel et al., 2001).
20
3.1 Description of the study area
The study was conducted on a constructed wetland located at Njoro campus of Egerton
University, Kenya. The Campus, is located about 25 Km south-west of Nakuru town in Nakuru
County. It stands on about 1580 hectares of land and lies within the River Njoro watershed at an
altitude of about 2238 m above sea level. This is an agricultural area characterized by bimodal
precipitation pattern ranging between 760 and 1270 mm per annum (Appendix 1). The long rains
fall from March – May while the short rains occur in September - November. The average
maximum and minimum temperatures range from 18 to 22 C and 5 to 8 C, respectively. The
soils are pre-dominantly andosols with a pH of 6.0 to 6.5 (Gogo et al., 2012).
With a population of about 18,000 people, the campus generates about 800 m3 /day of
wastewater which is treated within the University using waste stabilization ponds (lagoons) that
comprise of two anaerobic ponds, two facultative ponds and two maturation ponds. The
wastewater treatment system was designed to handle wastewater generated by a smaller
population when the institution was still an agricultural college. Deterioration of effluent quality
(Murage, 2003 and Mokaya et al., 2004) necessitated the construction of the wetland in series
with the waste stabilization ponds and next to the banks of River Njoro.
Consequently, through Maji na Ufanisi (an NGO), the GEF-SGP funded the development
of a constructed wetland to purify the wastewater from the WSPs in 2007. The wetland was
designed to purify about 100 m3/day of wastewater. Entering the wetland from the WSPs the
water runs through a gravel bed into three cells planted with water loving (hydrophilic) plants
which act as a biological filters before the water is finally discharged into River Njoro. The
wastewater retention time in the wetland is estimated to be 10 to 14 days. The wetland was also
supposed to serve as biodiversity conservancy, a nature trail and a recreation park and thereby
offer eco-tourism activities. The sedimentation/gravel bed is rectangular in shape and measures
approximately 50×21 m with alternating wall embankments within, which ensures that the
wastewater flow is slowed down before percolating through a section (about 100 m2) filled with
graded ballast before discharging into the wetland cells whose areas are shown in Table 2.
21
Table 2: Main Components of the Constructed Wetland and their sizes
Section of the Constructed Wetland Maximum Possible Area (m2)
Sedimentation/gravel bed 1050
22
Source: Google Map (Modified by G. Maina, Environmental Science Dept., Egerton University)
Figure 2: Map of the study area showing the location of the constructed wetland at Egerton
University in, (A) Kenya, (B) Nakuru County/Njoro Division and (C) Sampling points
23
3.2 Sampling points
Six sampling points (S1-S6) were chosen along the wetland system as shown on Figure 2.
Sampling point S1 was located at the inlet to the wetland (discharge point from the WSPs) and
was therefore not influenced by the wetland. S2 represents the sedimentation/gravel bed and the
dominant aquatic plant species was the Pistia stratiotes. Other plant species included Typha sp.,
Cyperus alopecuroides and Scirpus lacustris (Plate 5). S3 is wetland cell 1 and the dominant
aquatic plant was Eichhornia crassipes. Other plants were Cyperus alopecuroides, Scirpus
lacustris, Typha sp. and Cyperus papyrus (Plate 6). S4 is wetland cell 2 which was almost
completely covered by Water hyacinth with a few tufts of Cyperus alopecuroides on the fringes
(Plate 7). S5 represents Wetland cell 3 which was basically an open pond with scanty vegetation,
mainly Cyperus alopecuroides at the edges (Plate 8) while Point S6 was the outlet of wetland 3
where effluent discharged into River Njoro (Plate 9).
Plate 5: Sedimentation/gravel bed (S2) (Photo by Mwanyika F.T 2013)
24
Plate 6: Wetland cell 1(S3) (Photo by Mwanyika F.T 2013)
Plate 7 : Wetland cell 2(S4) (Photo by Mwanyika F.T 2013)
25
Plate 8 : Wetland cell 3(S5) (Photo by Mwanyika F.T 2013)
Plate 9: River Njoro at the wastewater discharge point (Photo by Mwanyika F.T 2013)
26
3.3 Sample collection
Sampling was carried out every four weeks over a period of six months (August 2013 to January
2014).
3.3.1 Collection of water samples
Water samples were collected in one litre plastic bottles which were pre-cleaned with
detergent, rinsed with tap water, soaked in 1M HNO3 overnight, rinsed with deionized water and
finally with the water samples from respective sampling. Three one litre samples of water were
collected from the sampling points between 9:00 am – 12:00 noon during all sampling occasions.
The water samples were labelled and acidified to pH < 2 using concentrated nitric acid (APHA,
2005), then transported to the laboratory for analysis. During each sampling session the
following parameters were determined in situ; water temperature, pH, DO and electrical
conductivity which were measured using a calibrated electrochemical analyzer (Jenway 3405)
fitted with probes for each parameter.
3.3.2 Collection of plant samples
For each sampling point, six plants were randomly collected and pooled together to form
a composite sample for each species. The whole shoot and root of each species were collected.
Freshly collected plant samples were stored separately in labelled polythene bags and transported
to the laboratory for processing.
3.3.3 Collection of sediment samples
Sediment samples were collected in triplicate from the sedimentation/gravel bed and the
three wetland cells respectively using a cylindrical core sampler inserted up to a depth of 15 cm.
The samples were then placed in clean labelled plastic bags and transported to the laboratory for
processing and heavy metals determination.
3.4 Samples preparation and analysis
3.4.1 Heavy metals analysis in wastewater samples
Duplicate samples of 30 mL each of the water samples were measured into 100 mL
conical flasks followed by 10 mL of Hydrochloric acid : water (1:1) and 2 mL Nitric acid: water
(1:1) solutions. The samples were digested on a hot plate until the volume reduced to about 25
mL. The samples were cooled to room temperature, filtered and diluted up to 50 mL in
27
volumetric flasks using deionized water (APHA, 2005). The samples were then analysed for
metals using Flame Atomic Absorption spectrophotometer. The Direct Air-Acetylene Flame
Method (APHA, 2005) was used to analyse the samples for Pb, Cu, Cd and Zn using a Flame
Atomic Absorption Spectrophotometer (model S11 from Thermo Jarell Ash Cooperation of
Waltham, MA, USA.). Standard solutions were serially diluted from stocks of 1000 mg/L
concentrations of each metal.
3.4.2 Heavy metals analysis in plant samples
Plant samples were washed with tap water then rinsed with deionized water to remove
soil particles and any aerial particulate deposition. They were allowed to drain and then
transferred to a regulated oven set at 60oC and kept there until much of the water had evaporated.
The temperature was increased to 80oC and the samples kept at this temperature until they
became crispy ready for grinding. They were ground using a hammer mill and sieved using a 0.5
mm mesh size sieve (Radojevi and Bashkin, 1999), then finally packed in labelled polythene
bags. Duplicate samples of 0.5 gram each of the ground plant material were weighed and
transferred into 100mL conical flasks followed by 10 mL of Hydrochloric acid: water (1:1) and 2
mL Nitric acid: water (1:1) solutions. The samples were digested on a hot plate at until the
volume reduced to about 5 mL. The samples were cooled to room temperature, filtered and
diluted up to 50 mL in volumetric flasks using deionized water. The samples were then analysed
for metals using the Flame Atomic Absorption spectrophotometer as described in sections 3.4.1.
3.4.3 Heavy metals analysis in sediment samples
Sediment samples were placed in evaporating dishes and oven dried at 105°C for 24
hours, followed by grinding and sieving through 0.5 mm sieve (Radojevi and Bashkin, 1999).
Duplicate samples of one gram each of dry sediment samples were weighed and placed in 100
mL conical flasks followed by 10 mL of Hydrochloric acid: water (1:1) and 2 mL Nitric acid:
water (1:1) solutions. The mixtures were digested on a hot plate until the volume reduced to
about 5 mL. The samples were cooled to room temperature, filtered and diluted up to 50 mL in
volumetric flasks using deionized water. The samples were then be analysed for metals using the
Flame Atomic Absorption spectrophotometer as described in sections 3.4.1 and 3.4.2.
28
3.5 Data analysis
Minitab statistical software was used to manage the data. The means for various
parameters were calculated and comparisons of variables performed using one-way analysis of
variance (ANOVA). The difference among means was tested using Tukey’s Honesty Significant
Different test (HSD test). The relationships between physico-chemical variables of water and
heavy metals concentrations in the wetland were determined using Pearson’s correlation
analysis. All the tests were carried out at 95% significance level.
The heavy metals removal efficiency of the wetland was determined by considering
influent and effluent concentrations of the metals and calculations were done using equation (1),
C1 - C2
x 100% removal =Metal (1)(Leta et al., 2004)
Where, C1 is the concentration of the metal in the influent and C2 is the concentration of the
metal in the effluent (units for metal concentrations are given in mg/L)
29
Values recorded for water temperature, conductivity, dissolved oxygen and pH during the
study period are presented in Table 3. Mean water temperature ranged between 17.3 ± 0.31 - 18.6
± 0.24 °C and showed significant difference across the wetland (ANOVA: F = 3.65, p = 0.029).
The lowest water temperature of 14.8 °C was recorded in S4 while the highest was 20.7 °C in S1.
There was a significant difference in conductivity along the wetland cells (ANOVA: F = 5.54, p
= 0.005). The highest concentration of dissolved oxygen was recorded in S5 (12.0 mg/L) while
the lowest concentration in the wetland was 0.1 mg/L, in S4. However, there was no significant
difference in oxygen concentration across the wetland cells (ANOVA: F = 1.48, p = 0.233). The
highest pH recorded was in S1 (8.5) while the lowest was in S4 (6.7).
Table 3: Mean values and ranges of physico-chemical parameters at different sites in the
wetland (values in parenthesis are the ranges).
Site Temp (oC)
4.2 Heavy metals concentration in the wetland.
4.2.1 Heavy metals concentration in water samples
Results for metal concentrations as determined in water samples are summarized in Table
4 and their variations along the constructed wetland wastewater treatment pathway are shown in
Figure 3. The highest mean lead concentration (1.25 ± 0.75 mg/L) was in S1 and lowest in S6
(0.07 ± 0.07 mg/L). There was no significant difference in lead concentration across the wetland
(ANOVA: F = 0.54, p = 0.705). Similarly, highest copper concentration (1.09 ± 0.49 mg/L) was
in S1 and lowest in S6 (0.32 ± 0.11 mg/L). There was no significant difference in copper
concentration across the wetland (ANOVA: F = 1.27, p = 0.282). The highest zinc concentration
(0.15 ± 0.11 mg/L) was in S1 and was not detected at S6. There was significant difference in
zinc concentration across the wetland (ANOVA: F = 10.69, p = 0.000). Lead concentration was
highest in S1 then dropped almost by half in S2, rose again in S3 then reduced consistently up to
S6. Although copper metal followed a similar trend as lead up to S3, its concentration continued
to rise up to S5 before dropping in S6. Zinc concentration was also highest in S1 but reduced in
subsequent sites until it was not detected in S6. Cadmium was below detectable limit of 0.01
mg/L in all samples in the wetland.
Table 4: Mean values of selected metals concentration in water samples at different sites in
the wetland (Data presented as Mean ± SE).
Site Pb (mg/L) Cu (mg/L) Zn (mg/L)
S1 1.25 ± 0.75 1.09 ± 0.49 0.15 ± 0.11
S2 0.69 ± 0.85 0.79 ± 0.40 0.09 ± 0.08
S3 0.97 ± 0.85 0.87 ± 0.56 0.08 ± 0.08
S4 0.83 ± 0.87 0.96 ± 0.60 0.05 ± 0.06
S5 0.69 ± 0.85 0.99 ± 0.65 0.03 ± 0.04
S6 0.07 ± 0.07 0.32 ± 0.11 ND*
NEMA MPL+ 0.1 mg/L 1.0 mg/L 0.5 mg/L
ANOVA F = 0.54
31
Figure 3: Variation of metal concentrations along the wetland profile. Vertical bars: ± SE
4.2.2 Heavy metals concentration in sediment samples
The concentrations of heavy metals in sediments for each sampling site are shown in
Table 5. The highest concentration of lead was in S5 (49.99 ± 0.01 mg/Kg) and the lowest in S4
(25 ± 27.38 mg/ Kg). There was no significant difference in lead concentration along the wetland
(ANOVA: F=1.75, p = 0.159). Copper concentration was highest in S2 (42.14 ± 18.63 mg/ Kg)
and reduced gradually along the wetland cells to a low of 27.97 ± 16.55 mg/ Kg in S5. There was
no significant difference in copper concentration along the wastewater flow path in the
constructed wetland (ANOVA: F=0.97, p = 0.408). Zinc concentration was highest in S3 (116.16
± 13.32 mg/ Kg) and lowest in S2 (84.21± 22.19 mg/ Kg). There was significant difference in
zinc concentration across the wetland cells (ANOVA: F = 29.61, p = 0.000).
-0.2
0.3
0.8
1.3
1.8
2.3
C o
n ce
n tr
at io
n ce
n tr
at io
n ce
n tr
at io
L)
Sites
Zn
32
Table 5: Mean values of selected metals concentrations in sediment samples at different
sites in the wetland (Data presented as Mean ± SE).
Site Pb (mg/Kg) Cu (mg/Kg) Zn (mg/Kg)
S2 41.64 ± 20.40 42.14 ± 18.63 84.21 ± 22.19
S3 41.66 ± 20.41 39.50 ± 23.00 116.16 ± 13.32
S4 25.00 ± 27.38 31.73 ± 12.96 109.77 ± 13.69
S5 49.99 ± 0.01 27.97 ± 16.55 95.38 ± 12.05
ANOVA F=1.75
4.2.3 Heavy metals concentration in plant samples
The results of the metal concentrations as determined in plants are summarized in Table
6. P. Stratiotes, found in S2 only, contained 84.37 ± 21.93 and 25.67 ± 10.16 mg/Kg copper and
zinc respectively, but lead was below detection limit. C. alopecuroides was found in all sites in
the wetland. Lead, copper and zinc were present in varying concentrations with exception of S5,
where lead was not detected. S. lacustris, present only in S2 and S3, showed the presence of
lead, copper and zinc. E. crassipes, in S3 and S4, was found to have lead, copper and zinc in
varying concentrations. Lead varied significantly in plants along the wetland cells (ANOVA: F =
2.58, p = 0.015). There was no significant difference in copper concentration in plants across the
wetland cells (ANOVA: F = 1.06, p = 0.399). Zinc concentration differed significantly in plants
along the wetland cells (ANOVA: F = 3.20, p = 0.003).
Table 6: Mean values of metals concentration in plant samples at different sites in the
wetland (Data presented as Mean ± SE).
Site Plant(s) Pb (mg/Kg) Cu (mg/Kg) Zn (mg/Kg)
S2
C. alopecuroides 19.99 ± 42.15 82.00 ± 18.36 13.80 ± 7.93
S. lacustris 5.00 ± 15.80 68.68 ± 29.02 26.07 ± 11.25
S3 E. crassipes 40.00 ± 51.64 71.39 ± 30.76 27.44 ± 13.42
C. alopecuroides 44.99 ± 49.71 86.98 ± 35.34 26.87 ± 13.48
S. lacustris 20.00 ± 42.16 95.31 ± 35.49 18.25 ± 7.54
S4 E. crassipes 15.00 ± 33.74 85.61 ± 33.44 11.66 ± 10.15
C. alopecuroides 49.99 ± 52.70 62.99 ± 33.58 29.03 ± 17.32
S5 C. alopecuroides ND* 78.65 ± 37.90 19.26 ± 6.31
ANOVA F = 2.58
4.3: Relationship between physico-chemical parameters and heavy metals in the wetland.
The Pearson correlation coefficient analyses between physico-chemical parameters of wastewater and the selected heavy
metals in water, sediments and plants is presented in Tables 7 - 10.
Table 7: Correlation coefficients between temperature and heavy metals in water (w), sediment (s) and plants (p).
Temp Znp Znw Pbs Pbp Pbw Cuw Cup Cus Zns
Znp -0.024 1
Pbp 0.006 -0.059 -0.059 1.000 1
Pbw 0.075 0.001 0.010 0.273 0.275 1
Cuw -0.125 -0.116 -0.116 -0.063 -0.063 0.058 1
Cup 0.101 -0.042 -0.042 0.152 0.152 0.259 -0.090 1
Cus 0.080 -0.118 -0.118 0.320 0.320 0.299 0.043 0.394 1
Zns -0.311 -0.262 -0.262 -0.103 -0.103 -0.121 0.326 -0.329 0.111 1
r ≥ 0.5 indicates existence of a strong correlation between parameters.
r < 0.5 indicates existence of a weak correlation between parameters.
The results of the Pearson correlation analysis between temperature and heavy metals in the wetland (Table 7) show existence of weak
correlation between temperature and metals concentration in the wetland.
34
Table 8: Correlation coefficients between DO and heavy metals in water (w), sediment (s) and plants (p).
DO Znp Znw Pbs Pbp Pbw Cuw Cup Cus Zns
Znp 0.111 1
Pbp 0.169 -0.059 -0.059 1.000 1
Pbw 0.061 0.010 0.010 0.275 0.275 1
Cuw -0.032 -0.116 -0.116 -0.063 -0.063 0.058 1
Cup 0.026 -0.042 -0.042 0.152 0.152 0.259 -0.090 1
Cus -0.175 -0.118 -0.118 0.320 0.320 0.299 0.043 0.394 1
Zns -0.317 -0.262 -0.262 -0.103 -0.103 -0.122 0.326 -0.329 0.111 1
r ≥ 0.5 indicates existence of a strong correlation between parameters.
r < 0.5 indicates existence of a weak correlation between parameters.
The results of the Pearson correlation analysis between DO and heavy metals in the wetland (Table 8) show existence of weak
correlation between dissolved oxygen and metals concentration in the wetland.
35
Table 9: Correlation coefficients between pH and heavy metals in water (w), sediment (s) and plants (p).
pH Znp Znw Pbs Pbp Pbw Cuw Cup Cus Zns
Znp 0.158 1
Pbp 0.111 -0.059 -0.059 1.000 1
Pbw 0.077 0.010 0.010 0.275 0.275 1
Cuw -0.236 -0.116 -0.116 -0.063 -0.063 0.058 1
Cup 0.468* -0.042 -0.042 0.152 0.152 0.259 -0.090 1
Cus 0.175 -0.118 -0.118 0.320 0.320 0.299 0.043 0.394 1
Zns -0.552* -0.262 -0.262 -0.103 -0.103 -0.122 0.326 -0.329 0.111 1
Significant levels are indicated by * at p < 0. 05.
r ≥ 0.5 indicates existence of a strong correlation between parameters.
r < 0.5 indicates existence of a weak correlation between parameters.
The Pearson correlation analysis between pH and heavy metals in the wetland showed that copper in plants was positively correlated
to pH while zinc in sediments was negatively correlated to pH (Table 9). However pH showed weak correlations with metals in the
other sample matrices in the wetland.
36
Table 10: Correlation coefficients between EC and heavy metals in water (w), sediment (s) and plants (p).
EC Znp Znw Pbs Pbp Pbw Cuw Cup Cus Zns
Znp -0.019 1
Pbp 0.302 -0.059 -0.059 1.000 1
Pbw 0.081 0.010 0.010 0.275 0.275 1
Cuw -0.040 -0.116 -0.116 -0.063 -0.063 0.058 1
Cup 0.034 -0.042 -0.042 0.152 0.152 0.259 -0.090 1
Cus 0.066 -0.118 -0.118 0.320 0.320 0.299 0.043 0.394 1
Zns -0.029 -0.262 -0.262 -0.103 -0.103 -0.121 0.326 -0.329 0.111 1
r ≥ 0.5 indicates existence of a strong correlation between parameters.
r < 0.5 indicates existence of a weak correlation between parameters.
The results of the Pearson correlation analysis between EC and heavy metals in the wetland (Table 10) show existence of weak
correlation between electrical conductivity and metals concentration in the wetland.
37
4.4 Heavy metals removal efficiency in the wetland
The efficiency of heavy metals removal from the wastewater in the wetland is
summarized in Table 11.
Metal Influent concentration
Zinc (Zn) 0.15 ± 0.11 ND* 100
ND* - Not Detected
In this study, metals removal efficiencies of 94.4%, 70.64% and 100% for lead, copper
and zinc respectively were observed.
38
Physicochemical parameters play an important role in determining the water quality and
greatly influence the functioning of the wetland. The accumulation of metals from the
wastewater to the sediment and plants is dependent on a number of external environmental
factors such as pH, dissolved oxygen, electrical conductivity and the available surface area for
adsorption (Davies et al., 2006).
In this study, the highest temperature was recorded in S1 (18.6 ± 0.24 oC) and could be
attributed to the direct sunlight into the last lagoon of the waste stabilization pond system. The
mean temperature drop in S2, S3 and S4 may have resulted from the vegetation mats over the
wetland cells coupled with low flow rate in the system. The temperature rise in S5 may be
attributed to the fact that the wetland cell was basically an open pond with limited vegetation
cover. Generally, wetland water temperature is assumed to be close to mean daily air temperature
(Kadlec and Knight, 1996). Temperature is an essential factor in life processes, including
biological, chemical and physical processes that occur in the environment. It influences growth
of macrophytes, therefore, it also affects the accumulation effect of heavy metals. When the
temperature is high, the growth and the metabolism of aquatic plants are high, leading to an
improvement of metal accumulation.
The pH values in the wetland were found to be within the limits set by NEMA (6.8 – 8.5)
for effluent discharge into the environment. The highest pH value was recorded in S1 (8.5) while
the lowest was in S4 (6.7). Kadlec and Wallace, (2009) reported that pH does not fluctuate
significantly throughout the year in most natural and constructed wetlands. Decrease in pH may
be attributed to organic substances generated within a wetland through growth, death and
decomposition cycles of wetland vegetation that are a source of natural acidity (Kadlec and
Knight, 1996). pH affects the solution chemistry of the metals, the activity of functional groups
in biomass and competition of metallic ions (Li et al., 2005). Redox potential and pH of the
sediment–water system are the major factors that influence the mobility of heavy metals in
wetlands (Koretsky et al., 2008).
Electrical conductivity is a measure of the ability of water to conduct an electric current
and it is related to the amount of dissolved ions in the water. It is a good and rapid method to
39
measure the total dissolved ions and is directly related to the total solids in the water sample
(Singh et al., 2010). Electrical conductivity in the wetland ranged between 690 – 885 µS/cm.
Variation of EC in the wetland cells can be attributed to the physico-chemical processes
occurring there, which include removal of ions through sedimentation, precipitation, adsorption
and uptake by aquatic plants. Although NEMA (Kenya) has not set the maximum permissible
levels for conductivity in the discharge, the South African guideline for conductivity in effluent
discharged into the receiving water bodies is 250 μS/cm (Government Gazette, 1984). NEMA
should gazette maximum permitted levels for EC since it is a very important parameter in water
to be used for irrigation.
The mean concentration of dissolved oxygen at the inlet (S1) was 4.7 ± 0.82 mg/L but
decreased in S2, S3 and S4 before rising slightly in S5 then drop again in S6. Decrease of oxygen
levels, especially in S2 and S3, may have been due to high oxygen demand by decomposing
organic matter and in the wetland cells. Dissolved oxygen concentrations above 1.5 mg/L are
required for most treatment processes (Ding et al., 2012). In unpolluted water, dissolved oxygen
concentrations normally range between 8 and 10 mg/L and concentrations below 5 mg/L
adversely affect aquatic life (Rao, 2005). The near-total cover by plant mats over the wetland
cells; low water velocity and settling of dead plants at the bed of the cells impede uniform flow
of water thereby creating channels and dead-ends within the system.
4.5.2 Heavy metals concentration in water samples
Metals in wastewater are present in colloidal, particulate, and dissolved phases. However,
more than 90% of heavy metals load in aquatic systems are found on suspended particles and
sediments while only a small portion of free metal ions remain dissolved in water (Zahra et al.,
2014). The decrease in mean concentrations of the metals in S2 is mainly attributed to
sedimentation. However, the variation in concentrations of the metals in water across the wetland
may be attributed to the various processes taking place in the wetland. Short-circuiting and
preferential flow paths were observed in the wetland, as demonstrated by the distribution patterns
of metal concentrations in the water sampled from each site. The decomposition process of
higher organic matter may lead to the lower value of pH due to the production of humic acid
(Nobi et al., 2010), which in turn increase metal release back to the water column. The wetland
40
is located next to horticultural fields and application of fertilizers and herbicides/fungicides could
be contributing copper and zinc into the wetland. The constructed wetland showed removal
efficiencies of 94.4%, 70.64% and 100% for lead, copper and zinc respectively. Gituku et al.
(2015) reported removal efficiencies of 80.0%, 95.1% and 100% for Pb, Cu and Zn respectively
from a study on Nandi Hills tea estates constructed wetlands and the results of these wetlands
compare favourably.
4.5.3 Heavy metals concentration in sediment samples
Sediments form the most significant component of water ecosystems with respect to
metal accumulation in wetlands (Willow and Cohen, 2003). Sediment quality is an important
indicator of water pollution and has been used as a tool to assess the health status of aquatic
ecosystems and is therefore, an integral component for functioning of ecological integrity (Zahra
et al., 2014). In this study Lead, copper and Zinc were present in sediments at much higher
concentrations than those observed in water. The metals concentrations in sediments across the
wetland ranged between 25 – 50 mg/Kg for lead, 28 – 42 mg/Kg copper and 84 – 116 mg/Kg
zinc. The accumulation process involves complex chemical and physical mechanisms which
depend on the nature of the sediment matrix and the properties of the adsorbed compounds
(Machate et al., 1999). pH and redox potential are the most important variables that govern
heavy metal solubility (Delaune et al., 1997). The high levels of metals found in the sediments
indicate that the wetland is quite good at retaining these metals by capturing sediments and
allowing for sedimentation of suspended metals. Wetlands provide anaerobic conditions that
promote the growth of sulfate-reducing bacteria. Lead, copper, cadmium and zinc form highly
insoluble sulfide compounds in contact with low concentrations of hydrogen sulfide. Sulfide
minerals have been reported in a number of wetland sediments (Sobolewski, 1999). However the
metal ions may renter the water column should the conditions e.g. pH and redox potential change
in the sediments or when there is perturbation in the wetland.
4.5.4 Heavy metals concentration in plant samples
All the four plant species investigated exhibited some phytoremediation characteristics
towards the selected metals. The levels of detected metals varied widely between/within plants
species and sites. This may be attributed to the intervening variables such as water volume and
41
flow as well as the prevailing physico-chemical conditions in each site. Wetland plants may
indirectly facilitate removal processes by providing surfaces for active and diverse microbial
communities (Rossmann et al., 2012), and preventing re-suspension of sedimented pollutants.
They not only assimilate pollutants directly into their tissues, but they also catalyze purification
reactions by increasing the environment diversity in the root zone and supporting a variety of
chemical and biochemical reactions that improve purification. Macrophytes alter the redox
conditions, pH and organic matter content of sediments and so affect the chemical speciation and
mobility of metals (Jacob and Otte, 2003).
In some cases metals get adsorbed onto plaques of iron and manganese oxides that form
on the roots of wetland plants (Batty et al., 2000). Root plaques have been observed to contain a
variety of metals and metalloids including aluminium, arsenic, cadmium, chromium, mercury,
nickel, lead and zinc (Sundby et al., 1998). This means that if conditions at any site favour the
formation of these plaques on the roots and submerged stems, then high concentrations of metals
will be observed.
4.5.5 Relationship between physico-chemical parameters of water and heavy metals
concentrations in the wetland.
The Pearson correlation analysis between physicochemical parameters of water and
heavy metals concentration in the wetland generally showed weak correlations (Table 6 - 9).
However, pH was positively correlated to copper in plants and negatively correlated to zinc in
sediments (Table 8). This implies that as pH increased in the wetland, copper concentration also
increased in the plants while zinc concentrations reduced in sediments. Haiyan et al. (2013)
reported that the effect of pH on Zn and Cu in sediments was such that the release of the metals
increased under both acidic and alkaline conditions. This may possibly explain the decrease of
Zinc level in sediments as pH increase; and the availability of copper for plant uptake. The
physicochemical parameters investigated, to some extent, do influence the concentrations of
heavy metals in the compartments of the wetland.
42
Based on the study findings, the following conclusions were made:
1. The physico-chemical parameters of wastewater varied spatially along the constructed
wetland wastewater treatment pathway.
2. The co